Materials Science and Engineering C 31 (2011) 238–245

Contents lists available at ScienceDirect

Materials Science and Engineering C

j ourna l homepage: www.e lsev ie r.com/ locate /msec

Growth of nacre in abalone: Seasonal and feeding effects

M.I. Lopez ⁎, P.Y. Chen, J. McKittrick, M.A. MeyersUniversity of California, San Diego, La Jolla, California, USA

⁎ Corresponding author.E-mail address: mil032@ucsd.edu (M.I. Lopez).

0928-4931/$ – see front matter. Published by Elsevierdoi:10.1016/j.msec.2010.09.003

a b s t r a c t

a r t i c l e i n f o

Article history:Received 20 March 2010Received in revised form 16 July 2010Accepted 3 September 2010Available online 9 September 2010

Keywords:AbaloneNacreGrowthHaliotisChitinBiomineralization

The processes of aggregation of mineral and organic materials to the growing surfaces in red abalone (Haliotisrufescens) are analyzed. The flat pearl implantation method is used to observe the transient stages of calciumcarbonate deposition, the structure of the organic interlayer, and the steady-state growth of aragonite tiles.The morphology of the organic interlayer is characterized by scanning electron microscopy. These resultsenable a realistic depiction of the formation of the terraced cones that comprise the principal biomin-eralization mechanism in this gastropod. In all cases, the growth initiated through spherulites, followed by tileformation. The transient stage with spherulitic formation was shorter at higher temperature; this is indicativeof a greater activity of the animal at 21 °C. The growth rate in a normally fed gastropod was found to be highercompared with one provided with limited food. The effect of water temperature (seasonal) was alsoestablished, with growth proceeding faster in the summer (T ~21 °C) than in winter (15 °C). The structures ofthe organic interlayer and of the epithelium are revealed by scanning electron microscopy.

B.V.

Published by Elsevier B.V.

1. Introduction

Understanding the process in which living organisms control thegrowth of structured inorganic materials can inspire new and bettersynthetic materials [1–5]. Indeed, there have been recent successes insynthesizing a ceramic/polymer composite with outstanding tough-ness inspired by the structure of nacre in the abalone shell [6–9].

The growth of nacre is a well studied subject characterized bymany researchers [10–35]. In particular, the growth and structurerelationship has been studied in detail [16,20–24,31]. Results showthat aragonite crystals first radiate from nucleation sites forming aspherulitic pattern, and then, columnar aragonite crystals formpreferentially in the c direction (perpendicular to the growth surface).This morphology is then replaced by the aragonite tile pattern. Linet al. [31] examined the structure during a period of 1 to 6 weeks. Inthe third week, the columnar growth still dominated and by the sixthweek growth cones of the aragonite nacre became present. Further-more, the role of the organic layer in the growth of the abalonenacre has been studied by Belcher et al. [22,23,26], Zaremba et al. [27],Sarikaya et al. [28–30], Lin et al. [15,31], Meyers et al. [16,17], andBezares et al. [34,35], which has led to proposed mechanisms ofgrowth.

However, little attention has been paid to factors that affect thedevelopment of these transient phases. Changes in the feedingpatterns may limit the source of ions for mineral formation in theabalone shell. Moreover, changes in its environment, such as

temperature of the sea water, might affect the nucleation rate andgrowth rate of the transitory phases of calcium carbonate. Thus, theenvironment may play an important role in the mineral formation.Additionally, past studies suggest a large involvement of the mantleand epithelial cell layer to form the intricate structure of the growingfront of the shell. Calcium radioisotope movement studies on theoyster Crassostrea virginica show that movement of the 45Ca out ofthe mantle correlated with the amount of 45Ca deposited on the shellgrowth front. Additional mollusk ion transport studies on the isolatedmantle indicate ion movements from the mantle to the shell, whileother studies suggest that Ca2+ transport occurs by diffusion throughthis mantle [38]. However, this process is not fully understood andstudies of this soft tissue can give insights into this biomineralizationprocess.

This study intends to investigate the process of mineralizationfollowing periods of growth interruption, taking into considerationimportant environmental factors (access to food and temperature)and to employ high-magnification characterization techniques tobetter understand how the soft tissue (e.g. epithelium and organicmembrane) influences the mechanism of growth.

2. Experimental techniques

Two labeled red abalone (Haliotis rufescens) were held in a 45 literfish tank in an open water facility at the Scripps Institution ofOceanography. The tank had direct access to continuously circulatingsea water, providing a natural environment with steady pH. Animalswere fed giant kelp (Macrocystis pyrifera) at different schedules andthe mean temperature was controlled. Three experiments werecarried out, varying average temperature and feeding rate of the

Fig. 1. Glass slides (depicted by arrows) embedded in abalone shell.

239M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

animal (Table 1). The ‘flat pearl’ technique, first used in the US by theU.C. Santa Barbara group [20,21,25,27] and latter applied by Lin et al.[31], was utilized to extract specimens for growth observations.Circular glass slides (15 mm in diameter) were implanted in liveabalone for periods of 1–3 weeks and then extracted weekly forexamination. The mantle was pressed back (retracted) with a flatscalpel and the slides were glued to the growing edge of the animal(Fig. 1). The largest quantity of slides allowed by the size and surfaceof the animal was implanted on each abalone. In Fig. 1, six slides wereimplanted and are shown by arrows. Once securely adhered, themantle relocated itself over slides over the period of approximately24 h. At least one slide from each of the abalone was removed weeklyand prepared for scanning electron microscopy (SEM) and atomicforce microscopy (AFM) characterization.

For SEM preparation, the slides where air dried and sputter coatedwith gold-platinum. The specimens were observed both from top andcross-sectional perspectives in the environmental and in the highvacuum modes of the Phillips XL30 ESEM. The surface morphology ofabalone nacre was also examined by AFM (Veeco Scanning ProbeMicroscope) in ambient dry condition.

The interpallial layer of mantle was also characterized. A smallslice (~1 cm) cut was made on two sections of live abalone (Fig. 2).Then each section was re-sectioned into two parts. Samples were CO2

critical point dried and gold-platinum coated for observation in theESEM in high vacuum mode.

Sections from the growing edge of abalone shell were also cut,washed in deionized water and demineralized in 0.6 N HCl at 20 °C for1 week. Specimens were then dehydrated completely in a progressivemanner in ethanol and CO2 critical point dried so that the structurewas maintained. Sections were gold-platinum coated and observed inthe ESEM high vacuum mode.

3. Results and discussion

3.1. Characterization of growth surfaces

The investigation in warm water (21 °C) revealed that thearagonite tiles formed after only 1 week (Fig. 3a). Furthermore, thefolding organic layers which are approximately 300 nm thick (markedby arrows) can be observed. Conversely, in coldwater (15 °C) (Fig. 3b)or with food limitation (20 °C) (Fig. 3c), observations after 1 weekshowed only the slight start of the precursor aragonite spread acrossthe substrate and some of the deposited mineral transitioning tospherulitic aragonite. Slides from the second abalone were alsoobserved and confirmed the same results. Moreover, week 2demonstrated similar aragonite tiles (Fig. 4a) when the growth wasconducted in warm water (21 °C). In contrast, uniform spheruliticaragonite was observed in cold water (15 °C) (Fig. 4b) or under nofeeding conditions (20 °C) (Fig. 4c). Interestingly, the spheruliticaragonite observed when the animal was not fed tends to be lessradiated compared to the structure in colder water. After 3 weeks ofgrowth in warm water (21 °C), a uniform and high number of stakedaragonite tiles (terraced cones) were observed (Figs. 5a and b). It canbe noted from Fig. 5a that because the height of the terraced cones isthe same, an even terrain is formed. In addition, from this cross-sectional view, a continuous membrane formed by the organic layercan be observed. From Fig. 5b it can be noted that the top of theterraced cones appears to be of a consistent diameter (~400 nm). On

Table 1Experimental conditions for sequential growth.

Condition Temperature of water Feeding schedule

1 ~21 °C Regularity2 ~15 °C Regularity3 ~20 °C Not fed

the contrary, after 3 weeks of implantation, the tops of eachspherulitic bundle form a plateau for test at 15 °C (Fig. 5c) and testwithout feeding (20 °C) (Fig. 5d). In addition, the thin organicmembrane can be observed (Fig. 5c).

AFM confirmed all the features observed by SEM. Fig. 6 showsthe growth surface in 21 °C. These mineral projections (terracedcones) are approximately 2 μm high, depicting about four layers fromthe top of the cones. This corresponds to the thickness of the tiles,(~0.5 μm). One can also see in Fig. 6b, on the sides of the protrusion,which represent terraced cones, the organic interlayer in a tent-likeformation. This is similar to the configurations seen in Figs. 3, 4 and 5,in which a thin organic layer covers terraced cones and demonstratesthat the organic layer, in its fully hydrated condition, stretches underits own weight. On the other hand, it acquires substantial strengthwhen it is dry [17]. In contrast to this, Fig. 7 demonstrates that onlythe spherulite morphology is attained with growth in water at 15 °C.Some distortion exists as the AFM tip does not capture very well thelateral details.

It should be clarified that the rate of the transition from initialrandomly arranged CaCO3, a spherulitic transient phase, to finalaragonite tile growth reported here is not the growth rate of thenacre, which is also affected by temperature and food availability

Fig. 2. Abalone mantle pushed back revealing epithelium (depicted by arrow) prior toexcision.

Fig. 3. Sequential growth results 1 week after implantation. a) Growth at 21 °C withabalone regularly fed; b) growth at 15 °C with abalone regularly fed; c) growth at 20 °Cwithout food available.

Fig. 4. Sequential growth results 2 weeks after implantation. a) Cross-sectional view ofgrowth at 21 °C with abalone regularly fed; b) growth at 15 °C with abalone regularlyfed; c) growth at 20 °C without feeding.

240 M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

[36]. Sequential growth results discussed by Lin et al. [31] demon-strated that the shell growth required various transitory phases toreach the steady-state growth of aragonite tiles. Aragonite tilesformation was achieved after approximately 6 weeks of precursortransitory phases (this previous study was performed at 15 °C andthe animal was fed regularly). The growth surfaces (Figs. 3b and 4b)show dominating spherulitic pattern and columnar growth which is

comparable to the results for 3 to 4 weeks described by Lin et al.[31].

In contrast, when the temperature was warmer (21 °C), the tran-sitions occurred faster. At this temperature, the transitory phasescannot be observed as the steady-state growth of aragonite tiles isreached by week one. Additionally, when the animal was not fed, thetransitions occurred later. The columns observed in the limited food

Fig. 5. Sequential growth results after 3 weeks of implantation. a) Growth at 21 °C with abalone regularly fed (cross-section); b) detailed view of aragonite tile morphology;c) growth in 15 °C with abalone regularly fed; d) growth in 20 °C without food available.

241M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

conditions at 20 °C tend to be less radiated and the surface lesssmooth when compared with the growth surfaces attained at 15 °C. Itis believed that the predominant columnar growth of the aragonitemineral is interrupted by the deposition of thin organic intertile layer[17]. The lack of nutrients and lower temperature may reduce theproduction of the organic layer (chitin and proteins), which leads toan unimpeded rapid columnar growth instead of the steady-stategrowth of the aragonite tiles.

3.2. Epithelium observations

Inspection of the mantle reveals the secretory epithelium whichis in direct contact with the inner surface of the shell (Fig. 8). Thispart of the animal is the critical component in the mineralization ofthe shell, since it is only separated from the growing surfaces by thesmall extrapallial space. SEM inspection (Fig. 8a) shows a top sectionof the outer surface of the epithelium (labeled I) and an area wherethe top surface is scraped off (labeled II). The relatively smooth outersurface of the epithelium (Fig. 8b) suggests that the epitheliummechanically flattens the growing surface by sliding over the shell,producing a molding effect [31] analogous to a potter molding clay.From the scraped surface one can observe that an array of channelsexists inside the epithelium (detailed view of the channels shown inFig. 8c). Some of these channels show fibrils (Fig. 8d). Lin et al. [31]and Meyers et al. [17] proposed that ions are allowed to diffusethrough these channels. In addition, these channels may providesupport for the synthesis of chitin and its intermittent extrusiononto the growth front (Fig. 9).

3.3. Demineralized shell and organic layer

The sectioned and demineralized shell samples from the growingedge revealed areas of thin organic intertile layer (Fig. 10a) inarrays of stretched holes. This organic intertile layer is believedto be periodically deposited (every ~0.5 μm) by the epitheliumin the animal. It is composed of a thin biopolymer protein frameworksecreted by epithelial cells [37–41]. This organic layer is an importantcharacteristic and has been studied successively [15–17,34,35].Lin and Meyers [15] investigated the thicker regions (20 μm) oforganic layer that exist between the shell's mesolayers. These thicklayers are believed to be formed by seasonal fluctuations wherecalcification is interrupted. Subsequently, Meyers et al. [16] furthernoted the role of the organicmatrix (20–50 nm thick) interlayer in theformation of the CaCO3 aragonite matrix into 0.5 μm thick tiles.Moreover, Meyers et al. [17] showed further evidence of the chitinnetwork that forms the structural component of the intertile layerand characterized it by SEM, AFM, and nanoindentation. Further-more, Bezares et al. [34,35] described the structure of demineralizedtissue and examined its mechanical response. In addition to beinga key element in the excellent mechanical properties, it is also animportant component in regulating the growth of the aragonite.This layer slows down the growth of the aragonite in the rapid growth(c axis) direction.

These results are in agreement with the growth mechanism pro-posed byMeyers and co-workers [17,31]. The growth of the mineral isallowed to proceed through the orifices in the organic layer as thetransport of calcium and carbonate ions is permitted through the

Fig. 6. Atomic force microscopy of growth surface in 21 °C showing the aragonite tilegrowth. a) Top view; b) tridimensional view.

Fig. 7. Atomic force microscopy of growth surface in 15 °C showing the columnarstructure. a) Top view; b) tridimensional view.

242 M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

holes in the organic layer (Fig. 10b). Fig. 11 shows randomly orientedchitin macromolecule fibrils [1,17,33,42–44] considered to be thestructural component of the organic layer.

There are two hypotheses explaining the formation and growth ofthe tiles:

(a) Organic scaffold, into which Ca2+ and CO32− ions penetrate,

combine, and precipitate [e.g. 34,35].(b) Periodic deposition of organic layer with holes, retarding the

growth of aragonite crystals in the c direction (perpendicular tothe growth surface) [17,31].

The organic scaffold hypothesis requires intricate genetic engi-neering. On the other hand, the periodic chitin deposition hypothesisis directly regulated by the mantle. The results obtained by Lin et al.

[31], Meyers et al. [17], and here strongly support the periodicdeposition hypothesis, a mechanism well described by Schäffer et al.[25] and Belcher et al. [22,26]. Especially significant is the fact that thelayers between laterally adjacent tiles are much thinner than thehorizontal ones (parallel to the growth surface), as shown in Fig. 10b.Of importance also is the identification of chitin synthesis sites in thecavernous channels within the epithelium; it is proposed that they areextruded onto the growth surface by mechanical action from theabalone foot.

4. Conclusions

In this study, the growth process of abalone nacre under differentenvironmental conditions (water temperature and food availability)

Fig. 8. a) Sectioned epithelium; surface in contact with growing edge of shell depicting flat outer surface (I) and area where surface scraped off (II); b) detailed view of flat outersurface of epithelium. c) Array of channels within epithelium (channels depicted by arrows); d) fibrils within channels (marked with arrows).

243M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

is investigated by the flat pearl technique combined with SEM andAFM observations. Demineralization studies were also performed. Themajor findings are:

1. The rate of the transition from an initial random nucleation to thesteady-state growth of aragonite tiles in abalone nacre is greatlyaffected by water temperature and food availability. The aragonitetile growth is achieved faster at warmer water temperature (21 °C)compared with colder temperature (15 °C). The transition fromcolumnar to aragonite tile growth takes longer when the food

Fig. 9. Schematic depiction of hypothetical mechanism by which epitheli

supply is limited. It is proposed that the lack of nutrients and lowertemperature may reduce the production of the intertile organiclayer, which leads to columnar growth of mineral instead of thearagonite tiles.

2. An outer smooth surface and inner channels are observed in theepithelium. These channels provide a path for ion and chitintransport. The smooth outer surface is believed to mechanicallyflatten the growing surface of the shell.

3. The demineralized shell showed sections of the intertile layer andchitin network. This organic component is believed to influence

um generates chitin fibrils and ‘squeezes’ them onto growth surface.

Fig. 10. a) Thin intertile organic layer showing holes; b) proposedmechanism of growthof nacreous tiles by formation of mineral bridges as depicted by Meyers et al. [17];organic layer is permeable to calcium and carbonate ions which nourish lateral growthas periodic secretion and deposition of the organic intertile membranes restrict theirflux to the lateral growth surfaces. Arrows A designate organic interlayer imaged bySEM; arrow B designates lateral boundary of tile (adapted from Meyers et al. [15]).

244 M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

growth mechanism of aragonite tiles by retarding, periodically, thegrowth in the c axis direction.

4. The results presented here strongly support the mechanism ofperiodic chitin layer deposition, retarding the growth of aragonite inthe c direction and forming the organic intertile layers. Cavernouschannels within the epithelium are identified and proposed to besites for chitin synthesis.

Fig. 11. a) Demineralized shell revealing randomly oriented chitin fibrils from intertilelayers; b) schematic representation of organic intertile layer composed of chitin fibrils.c) Schematic structure of chitin.

Acknowledgements

This research is supported by the National Science FoundationGrant DMR 0510138 and UCSD Alliance for Graduate Education andthe Professoriate Fellowship. We thank Ryan Anderson and Chung-Ting Wei for their assistance at CaIit2 facility. Thanks go to EddieKisfauludy for maintenance of the open water facility and the feedingof abalone. Yen-Shan Lin provided aid in demineralization procedure.We thank Maribel Montero (CalIT2) for helping with AFM measure-ments. Assistance provided by Aruni Suwarnasarn with the scanningelectron microscopy is greatly acknowledged.

245M.I. Lopez et al. / Materials Science and Engineering C 31 (2011) 238–245

References

[1] M. Sarikaya, Microsc. Res. Tech. 27 (1994) 360.[2] A.V. Srinivasan, G.K. Haritos, F.L. Hedberg, Appl. Mech. Rev. 44 (1991) 463.[3] G. Mayer, Science 310 (2005) 1144.[4] C. Sanchez, H. Arribart, M.M. Giraud-Guille, Nat. Mater. 4 (2005) 277.[5] G. Mayer, M. Sarikaya, Exper. Mech. 42 (2002) 395.[6] M.E. Launey, E. Munch, D.H. Alsem, H.B. Barth, E. Saiz, A.P. Tomsia, R.O. Ritchie,

Acta Mater. 57 (2009) 2919.[7] E. Munch, M.E. Launey, D.H. Alsem, E. Saiz, A.P. Tomsia, R.O. Ritchie, Science 322

(2008) 1516.[8] S. Deville, E. Saiz, R.K. Nalla, A.P. Tomsia, Science 311 (2006) 515.[9] S. Deville, E. Saiz, A.P. Tomsia, Biomaterials 27 (2006) 5480.

[10] G. Bevelander, H. Nakahara, in: M. Omori, N. Watabe (Eds.), The Mechanisms ofBiomineralization in Animals and Plants, Tokai University Press, Tokyo, 1980,p. 19.

[11] K. Wada, Biomineralization 6 (1972) 84.[12] H. Nakahara, Venus 38 (1979) 205.[13] G. Bevelander, H. Nakahara, M. Kakei, Venus Jpn. J. Malac. 41 (1982) 33.[14] H. Nakahara, Calcification of gastropod nacre, in: P. Westbroek, E.W. De Jong

(Eds.), Biomineralization and Biological Metal Accumulation, D. Reidel PublishingCompany, Dordrecht, Holland, 1982, p. 225.

[15] A. Lin, M.A. Meyers, Mater. Sci. Eng. A 390 (2005) 27.[16] M.A. Meyers, A.Y.M. Lin, P.Y. Chen, J. Muyco, J. Mech. Beh. Bio. Mat. 1 (2008) 76.[17] M.A. Meyers, C.T. Lim, A. Li, B.R. Hairul Nizam, E.P.S. Tan, Y. Seki, J. McKittrick, Mat.

Sci. Eng. C. 29 (2009) 2398.[18] H. Nakahara, in: S. Suga, H. Nakahara (Eds.), Mechanisms and Phylogeny of

Mineralization in Biological Systems, Springer-Verlag, New York, 1991, p. 343.[19] H. Mutvei, in: M. Omori, N. Watabe (Eds.), The Mechanisms of Biomineralization

in Animals and Plants, Tokai Univ. Press, 1980, p. 49.[20] M. Fritz, A.M. Belcher, M. Radmacher, D.A. Walters, P.K. Hansma, G.D. Strucky, D.E.

Morse, Nature 371 (1994) 49.[21] M. Fritz, D.E. Morse, Col. Int. Sci. 3 (1998) 55.[22] A.M. Belcher, X.H. Wu, R.J. Christensen, P.K. Hansma, G.D. Stucky, D.E. Morse,

Nature 381 (1996) 56.

[23] A.M. Belcher, E.E. Gooch, in: E. Bauerlein (Ed.), Biomineralization: From Biology toBiotechnology and Medical Application, Wiley-Interscience, Germany, 2000,p. 221.

[24] X.Y. Shen, A.M. Belcher, P.K. Hansma, G.D. Stucky, D.E. Morse, J. Bio. Chem. 272(1997) 32472.

[25] T.E. Schäffer, C. Ionescu-Zanetti, R. Proksch, M. Fritz, D.A.Walters, N. Almqvist, C.M.Zaremba, A.M., Chem. Mater. 8 (1996) 679.

[26] A.M. Belcher, B.L. Smith, G.D. Stucky, D.E. Morse, P.K. Hansma, Chem. Mater. 9(1997) 1731.

[27] C.M. Zaremba, A.M. Belcher, M. Fritz, Y. Li, S. Mann, P.K. Hansma, D.E. Morse, J.S.Speck, G.D. Stucky, Chem. Mater. 8 (1996) 679.

[28] M. Sarikaya, K.E. Gunnison, M. Yasrebi, J.A. Aksay, Mater. Res. Soc. 174 (1990) 109.[29] M. Sarikaya, J.A. Aksay, in: S. Case (Ed.), Results and Problems in ‘Cell

Differentiation in Biopolymers’, Springer, Amsterdam, 1992, p. 1.[30] M. Sarikaya, J. Microscopy Res Techn. 427 (1994) 360.[31] A.Y.M. Lin, P.Y. Chen, M.A. Meyers, Acta Biomater. 4 (2008) 131.[32] H.D. Espinosa, J.E. Rim, F. Barthelat, M. Buehler, Prog. Mater. Sci. 54 (2009) 1059.[33] A.G. Checa, J.H.E. Cartwright, M.-G. Willingerc, Proc. Natl Acad. Sci. USA 106

(2009) 38.[34] J. Bezares, R.J. Asaro, M. Hawley, J. Struct. Biol. 163 (2008) 61.[35] J. Bezares, R.J. Asaro, M. Hawley, J. Struct. Biol. 170 (2010) 484.[36] A. Steinarsson, A.K. Imsland, Aquaculture 224 (2003) 353.[37] H.A. Lowenstam, S. Weiner, On Biomineralization, Oxford U Press, 1989.[38] K. Simkiss, K. Wilbur, Biomineralization, Cell Biology and Mineral Deposition,

Academic Press, 1989.[39] A.M. Belcher, E.E. Gooch, Protein components and inorganic structure in shell

nacre, Biomineralization, Wiley-Vch, New York, 2000, p. 221.[40] L. Addadi, D. Joester, F. Nudelman, S. Weiner, Chem. Eur. J. 12 (2006) 980.[41] M. Rousseau, E. Lopez, P. Stempflé, M. Brendlé, L. Franke, A. Guette, R. Naslain, X.

Bourrat, Biomaterials 26 (2005) 6254.[42] M.A. Crenshaw, H. Ristedt, in: N. Omori, N. Watabe (Eds.), The Mechanisms of

Biomineralization in Animals and Plants, Tokay U. Press, 1976.[43] S. Weiner, W. Traub, FEBS Lett. 111 (1980) 311.[44] S. Weiner, W. Traub, Phyl. Trans. Roy. Soc. B. 304 (1984) 421.